Dynamic Metasurface Antenna-Enabled Multicast Transmission With Finite-Alphabet Inputs

Dynamic metasurface antennas (DMAs) have emerged as a promising solution for alleviating the high power consumption, hardware cost, and physical size of multiple-input multiple-output (MIMO) architectures. This paper investigates a DMA-enabled multicast system with finite-alphabet inputs, considering the availability of either instantaneous or statistical channel state information (CSI). For the instantaneous CSI scenario, a multicast rate (MR) maximization problem is formulated, and an efficient two-stage equivalent precoder approximation (EPA)–based algorithm is proposed to jointly optimize the digital precoder and the DMA weight matrix. For the statistical CSI scenario, the ergodic MR (EMR) maximization problem is considered. To gain fundamental insights into the system design, asymptotic analyses of the EMR are conducted in the high signal-to-noise ratio (SNR) regime. The results reveal that the EMR converges to a finite constant as the SNR approaches infinity, with the convergence rate (CR) determined by the diversity order and the array gain. Motivated by these analytical insights, an array gain maximization problem is formulated and an EPA-based algorithm is proposed. Numerical results demonstrate the effectiveness of the proposed algorithms, validate the correctness of the derived high-SNR EMR analysis, and confirm the effectiveness of using DMA to enhance multicast transmission.